Dual energy CT arthrography in shoulder instability: successful iodine removal with virtual non-contrast images and accurate 3D reformats of the glenoid for assessment of bone loss

Christoph Stern,Magda Marcon,Andrea B Rosskopf,Samy Bouaicha,Karl Wieser,Reto Sutter

doi:10.1007/s00256-021-03916-3

Christoph Stern, Magda Marcon + Show 4 more

Open Access

https://doi.org/10.1007/s00256-021-03916-3

Copy DOI

Abstract

ObjectiveTo evaluate the image quality of dual energy CT (DECT) of the shoulder after arthrography and of virtual non-contrast (VNC) 3D reformats of the glenoid and to compare glenoid measurements on VNC 3D reformats and on 2D CTs.Materials and methodsDECT arthrography (80 kV/140 kV) was performed in 42 shoulders of 41 patients with instability using diluted iodinated contrast media (80 mg/ml). VNC images and VNC 3D reformats of the glenoid were calculated using image postprocessing. Dose parameters, CT values of intraarticular iodine and muscle, image contrast (iodine/muscle), and image quality (5-point scale: 1 = worst, 5 = best) were evaluated. Two independent readers assessed glenoid morphology and performed glenoid measurements on 2D and 3D images.ResultsCalculation of VNC images and VNC 3D reformats was successful in 42/42 shoulders (100%). The effective dose was mean 1.95 mSv (± 0.9 mSv). CT values of iodine and muscle were mean 1014.6 HU (± 235.8 HU) and 64.5 HU(± 8.6 HU), respectively, and image contrast was mean 950.2 HU (± 235.5 HU). Quality of cross-sectional images, VNC images, and VNC 3D reformats was rated good (median 4 (4–5), 4 (3–4), 4 (3–5), respectively). Detection of an osseous defect was equal on 2D and 3D images (13/42, P > 0.99) with no difference for measurement of the glenoid diameter with mean 28.3 mm (± 2.8 mm) vs. 28.4 mm (± 2.9 mm) (P = 0.5), width of the glenoid defect with 3.2 mm (± 2.1 mm) vs. 3.1 mm (± 2.3 mm) (P = 0.84), surface area with 638.5 mm2 (± 127 mm2) vs. 640.8 mm2 (± 129.5 mm2) (P = 0.47), and surface area of the defect with 46.6 mm2 (± 44.3 mm2) vs. 47.2 mm2 (± 48.0 mm2) (P = 0.73), respectively.ConclusionDECT shoulder arthrography is feasible and allows successful iodine removal with generation of VNC images and accurate VNC 3D reformats of the glenoid for assessment of bone loss.

Highlights

The 3D reformats of shoulder CT examinations generated with the volume rendering technique (VRT) are routinely used for assessment of the glenoid morphology and for quantitative measurements and are highly accurate
We showed that dual energy CT (DECT) shoulder arthrography is feasible for clinical routine examinations and allows successful iodine removal using material decomposition with generation of virtual non-contrast (VNC) images and accurate 3D VRT reformats of the glenoid for assessment of bone loss
Subhas et al showed diagnostic image quality for CT scans of human knee cadavers performed in single energy technique with 80 kV after arthrography using

Summary

Introduction

The 3D reformats of shoulder CT examinations generated with the volume rendering technique (VRT) are routinely used for assessment of the glenoid morphology and for quantitative measurements and are highly accurate. They are often used during preoperative planning in shoulder instability procedures, especially as the 3D visualization is preferred by the surgeons over the 2D CT reformats. For SECT scans after arthrography, calculation of 3D VRT reformats is practically useless, because intraarticular iodinated contrast material overlays bone and cannot be successfully separated. For CT arthrography, DECT has only been tested in vitro in cow femoral condyles [12] and porcine joint cadavers [13] and was applied to shoulder patients for evaluation of labral tears [14] and for successful bone-iodine differentiation using material decomposition but without generating virtual noncontrast (VNC) images [15] or glenoid evaluation [16]

Methods

Results

Conclusion